D. Moreau, ITPA-IOS Meeting, Kyoto, October 18-21, 2011 LEHIGH U N I V E R S I T Y First Integrated Magnetic and Kinetic Control for AT Scenarios on DIII-D.

Slides:



Advertisements
Similar presentations
EXTENDED MHD SIMULATIONS: VISION AND STATUS D. D. Schnack and the NIMROD and M3D Teams Center for Extended Magnetohydrodynamic Modeling PSACI/SciDAC.
Advertisements

November 3-5, 2003Feedback Workshop, Austin NORMAL MODE APPROACH TO MODELING OF FEEDBACK STABILIZATION OF THE RESISTIVE WALL MODE By M.S. Chu(GA), M.S.
CHE 185 – PROCESS CONTROL AND DYNAMICS
Chapter 4: Basic Properties of Feedback
Controller Design, Tuning. Must be stable. Provide good disturbance rejection---minimizing the effects of disturbance. Have good set-point tracking---Rapid,
Stability, Transport, and Conrol for the discussion Y. Miura IEA/LT Workshop (W59) combined with DOE/JAERI Technical Planning of Tokamak Experiments (FP1-2)
Performance oriented anti-windup for a class of neural network controlled systems G. Herrmann M. C. Turner and I. Postlethwaite Control and Instrumentation.
Robust control Saba Rezvanian Fall-Winter 88.
Exponential Tracking Control of Hydraulic Proportional Directional Valve and Cylinder via Integrator Backstepping J. Chen†, W. E. Dixon‡, J. R. Wagner†,
280 SYSTEM IDENTIFICATION The System Identification Problem is to estimate a model of a system based on input-output data. Basic Configuration continuous.
NORM BASED APPROACHES FOR AUTOMATIC TUNING OF MODEL BASED PREDICTIVE CONTROL Pastora Vega, Mario Francisco, Eladio Sanz University of Salamanca – Spain.
FOM-Institute for Plasma Physics Rijnhuizen Association Euratom-FOM Tony Donné Introduction to fusion control.
November 21, 2005 Center for Hybrid and Embedded Software Systems Engine Hybrid Model A mean value model of the engine.
HEAT TRANSPORT andCONFINEMENTin EXTRAP T2R L. Frassinetti, P.R. Brunsell, M. Cecconello, S. Menmuir and J.R. Drake.
Recent Results of Configuration Studies L. P. Ku Princeton Plasma Physics Laboratory ARIES-CS Project Meeting, November 17, 2005 UCSD, San Diego, CA.
CH 1 Introduction Prof. Ming-Shaung Ju Dept. of Mechanical Engineering NCKU.
Fast imaging of global eigenmodes in the H-1 heliac ABSTRACT We report a study of coherent plasma instabilities in the H-1 plasma using a synchronous gated.
Y. Sakamoto JAEA Japan-US Workshop on Fusion Power Plants and Related Technologies with participations from China and Korea February 26-28, 2013 at Kyoto.
Recent Results from the STOR-M Tokamak A.Hirose, M. Dreval, S. Elgriw, O. Mitarai(1), A. Pant, M. Peng(2), D. Rohraff, A.K. Singh(3), D. Trembach, C. Xiao.
Advanced Tokamak Plasmas and the Fusion Ignition Research Experiment Charles Kessel Princeton Plasma Physics Laboratory Spring APS, Philadelphia, 4/5/2003.
SIMULATION OF A HIGH-  DISRUPTION IN DIII-D SHOT #87009 S. E. Kruger and D. D. Schnack Science Applications International Corp. San Diego, CA USA.
Nonlinear Frequency Chirping of Alfven Eigenmode in Toroidal Plasmas Huasen Zhang 1,2 1 Fusion Simulation Center, Peking University, Beijing , China.
Chapter 6 Control Using Wireless Throttling Valves.
Excitation of ion temperature gradient and trapped electron modes in HL-2A tokamak The 3 th Annual Workshop on Fusion Simulation and Theory, Hefei, March.
A Singular Value Decomposition Method For Inverting Line Integrated Electron Density Measurements in Magnetically Confined Plasma Christopher Carey, The.
1 Humphreys/55 th APS-DPP/October 2012 D.A. Humphreys 1, G.L. Jackson 1, R. Hawryluk 2, E. Kolemen 2, D. Moreau 3, A. Pironti 4, G. Raupp 5, O. Sauter.
Current Drive for FIRE AT-Mode T.K. Mau University of California, San Diego Workshop on Physics Issues for FIRE May 1-3, 2000 Princeton Plasma Physics.
現代制御理論を用いた核融合プ ラズマ制御シミュレーション 三善 悠矢 東京大学大学院新領域創成科学研究科.
OPERATIONAL SCENARIO of KTM Dokuka V.N., Khayrutdinov R.R. TRINITI, Russia O u t l i n e Goal of the work The DINA code capabilities Formulation of the.
PID Controller Design and
Low Level Control. Control System Components The main components of a control system are The plant, or the process that is being controlled The controller,
APS DPP 2006 October Adaptive Extremum Seeking Control of ECCD for NTM Stabilization L. Luo 1, J. Woodby 1, E. Schuster 1 F. D. Halpern 2, G.
Workshop on MHD Control 2008 – O.Katsuro-Hopkins Computational analysis of advanced control methods applied to RWM control in tokamaks Oksana N. Katsuro-Hopkins.
OPERATIONAL SCENARIO of KTM Dokuka V.N., Khayrutdinov R.R. TRINITI, Russia O u t l i n e Goal of the work The DINA code capabilities Formulation of the.
Feedback Control system
M. Onofri, F. Malara, P. Veltri Compressible magnetohydrodynamics simulations of the RFP with anisotropic thermal conductivity Dipartimento di Fisica,
Review: Neural Network Control of Robot Manipulators; Frank L. Lewis; 1996.
CARE / ELAN / EUROTeV Feedback Loop on a large scale quadrupole prototype Laurent Brunetti* Jacques Lottin**
Control systems KON-C2004 Mechatronics Basics Tapio Lantela, Nov 5th, 2015.
Lecture 25: Implementation Complicating factors Control design without a model Implementation of control algorithms ME 431, Lecture 25.
STUDIES OF NONLINEAR RESISTIVE AND EXTENDED MHD IN ADVANCED TOKAMAKS USING THE NIMROD CODE D. D. Schnack*, T. A. Gianakon**, S. E. Kruger*, and A. Tarditi*
L. Zanotto – 10 February 2011 – RFX-mod programme workshop TF2: Active control of RFP and tokamak plasmas Report on experimental proposals L. Zanotto,
TITLE Stuart R.Hudson, D.A.Monticello, A.H.Reiman, D.J.Strickler, S.P.Hirshman, L-P. Ku, E.Lazarus, A.Brooks, M.C.Zarnstorff, A.H.Boozer, G-Y. Fu and G.H.Neilson.
Chapter 4 A First Analysis of Feedback Feedback Control A Feedback Control seeks to bring the measured quantity to its desired value or set-point (also.
Steady State Discharge Modeling for KSTAR C. Kessel Princeton Plasma Physics Laboratory US-Korea Workshop - KSTAR Collaborations, 5/19-20/2004.
1 Feature of Energy Transport in NSTX plasma Siye Ding under instruction of Stanley Kaye 05/04/09.
045-05/rs PERSISTENT SURVEILLANCE FOR PIPELINE PROTECTION AND THREAT INTERDICTION Taming The Physics For Commercial Fusion Power Plants ARIES Team Meeting.
Disturbance rejection control method
Diagnostics and Optimization Procedures for Beamline Control at BESSY A. Balzer, P. Bischoff, R. Follath, D. Herrendörfer, G. Reichardt, P. Stange.
Plasma MHD Activity Observations via Magnetic Diagnostics Magnetic islands, statistical methods, magnetic diagnostics, tokamak operation.
1PAC-37, Plasma control algorithm development on NSTX-U using TRANSP, M.D. Boyer, 1/26/2016 Dan Boyer for the Integrated Scenarios science group Plasma.
NSTX Meeting name – abbreviated presentation title, abbreviated author name (??/??/20??) Goals of NSTX Advanced Scenario and Control TSG Study, implement,
1 ASTRA simulations of ITER long pulse scenarios V. Leonov ASTRA simulations of ITER long pulse scenarios V. Leonov Kurchatov Institute, Moscow, Russia.
APS DPP 2006 October Dependence of NTM Stabilization on Location of Current Drive Relative to Island J. Woodby 1 L. Luo 1, E. Schuster 1 F. D.
1 E. Kolemen / IAEA / October 2012 Egemen Kolemen 1, A.S. Welander 2, R.J. La Haye 2, N.W. Eidietis 2, D.A. Humphreys 2, J. Lohr 2, V. Noraky 2, B.G. Penaflor.
TH/7-1Multi-phase Simulation of Alfvén Eigenmodes and Fast Ion Distribution Flattening in DIII-D Experiment Y. Todo (NIFS, SOKENDAI) M. A. Van Zeeland.
Energetic ion excited long-lasting “sword” modes in tokamak plasmas with low magnetic shear Speaker:RuiBin Zhang Advisor:Xiaogang Wang School of Physics,
O. Sauter “Robust” NTM Control: The AMN-system O. Sauter for the TCV and AUG teams Ecole Polytechnique Fédérale de Lausanne (EPFL) Centre de Recherches.
Page 1 ITPA IOS Kyoto, Oct 2011 A new approach to plasma profile control in ITER S.H. Kim 1 and J.B. Lister 2 1 ITER Organization, St Paul lez Durance,
Presentation at NI Day April 2010 Lillestrøm, Norway
8th IAEA Technical Meeting on
Overall Objectives of Model Predictive Control
Linear Control Systems
Basic Design of PID Controller
Influence of energetic ions on neoclassical tearing modes
Dynamical Systems Basics
Hafez Sarkawi (D1) Control System Theory Lab
New Results for Plasma and Coil Configuration Studies
20th IAEA Fusion Energy Conference,
PID Controller Design and
Presentation transcript:

D. Moreau, ITPA-IOS Meeting, Kyoto, October 18-21, 2011 LEHIGH U N I V E R S I T Y First Integrated Magnetic and Kinetic Control for AT Scenarios on DIII-D D. Moreau CEA-DSM-IRFM, Cadarache, St-Paul-lez-Durance Cedex, France E. Schuster, J. Barton, D. Boyer, W. Shi, W. Wehner Lehigh University, Bethlehem, PA 18015, USA M. Walker, J. Ferron, D. Humphreys, A. Hyatt, B. Johnson, B. Penaflor, F. Turco, and many DIII-D collaborators General Atomics, San Diego, CA 92186, USA 1 ITPA-IOS Joint Experiment # 6.1

D. Moreau, ITPA-IOS Meeting, Kyoto, October 18-21, 2011 LEHIGH U N I V E R S I T Y Outline ARTAEMIS model structure and integrated control Control of the internal poloidal flux profile,  (x) Integrated control of  (x) and  N Lehigh University controller and  (x) control Proposal for 2012 and beyond 2

D. Moreau, ITPA-IOS Meeting, Kyoto, October 18-21, 2011 LEHIGH U N I V E R S I T Y The ARTAEMIS (grey-box) model-based approach D. Moreau et al., Nucl. Fusion 48 (2008) ARTAEMIS is a set of algorithms that use singular perturbation methods : (i)a semi-empirical system identification method (ii)a model-based, 2-time-scale, control algorithm for magneto-kinetic plasma state What could a minimal state space model look like ? Are there natural state variables and input variables ? How are they coupled ? etc … Generic structure of linearized flux-averaged plasma transport equations :

D. Moreau, ITPA-IOS Meeting, Kyoto, October 18-21, 2011 LEHIGH U N I V E R S I T Y The ARTAEMIS controller design and parameters for combined  (x) and  N control 4 Singular perturbation analysis  Near-optimal control = Optimal control up to O(  2 ) The dynamics minimizes using a given set of SVD modes (controller order) minimizes The slow proportional + integral feedback tracks a steady state that The fast proportional feedback loop maintains the kinetic variables, e. g.  N, on a trajectory which is consistent with the slow magnetic state evolution,  (x, t). given weight matrices, Q and R, with X = controlled variables and u = actuators to control simultaneously  (x) and  N minimizes

D. Moreau, ITPA-IOS Meeting, Kyoto, October 18-21, 2011 LEHIGH U N I V E R S I T Y Control of the internal poloidal flux profile :  (x) =  (x) –  boundary 5 The controller minimizes with actuator constraints and optimal gain matrices that depend on controller parameters : 4 actuators = NB-Co, NB-Bal, ECCD (5 gyros), Vsurf (NB 210R unavailable on 09/13) R-matrix : actuator weight fixed by considering actuator headroom (MW & Volts ?) Q-matrix : same weight on 9 different controlled radii (x = 0.1, …, 0.9) Controller order = 2 (proportional + integral control) Weight on integral control in the Q-matrix = 4, 10, 25, respectively, on the 3 examples below : # : IntWeight = 4# : IntWeight = 10# : IntWeight = 25

D. Moreau, ITPA-IOS Meeting, Kyoto, October 18-21, 2011 LEHIGH U N I V E R S I T Y Control of the poloidal flux profile (x = 0.1, 0.2, … 0.9) 6 # : IntWeight = 4# : IntWeight = 10# : IntWeight = 25 Control : 3.5 s - 6 sControl : 3.5 s - 5 sControl : 2.5 s - 6 s x = 0.1 x = 0.2 x = 0.3 x = 0.4 x = 0.5 x = 0.6 x = 0.7 x = 0.8 x = 0.9

D. Moreau, ITPA-IOS Meeting, Kyoto, October 18-21, 2011 LEHIGH U N I V E R S I T Y Control of the poloidal flux profile  t = 2.5 s, 4 s, 6 s 7 # : IntWeight = 4# : IntWeight = 25 SlowFast (overshoot)

D. Moreau, ITPA-IOS Meeting, Kyoto, October 18-21, 2011 LEHIGH U N I V E R S I T Y with actuator constraints and optimal gain matrices that depend on controller parameters : 5 actuators = NB-Co, NB-Bal, NB-Cnt, ECCD (6 gyros), Vsurf R-matrix : actuator weights fixed by considering actuator headroom (MW & Volts ?) Q-matrix : same weight on 9 controlled radii for  (x), x=0.1, 0.2, … 0.9 Weight on  N control : = 0.3 Controller order = 2 or 3 (proportional + integral control) Weight on integral control in the Q-matrix = 25 and 10, respectively, in the 2 examples below : Actuator saturation Simultaneous control of the  (x) profile and  N 5 actuators : NB-co, NB-bal, NB-cnt, ECCD, Vsurf 8 The controller minimizes # Control order = 2 IntWeight = 25 # Control order = 3 IntWeight = 10 MHD

D. Moreau, ITPA-IOS Meeting, Kyoto, October 18-21, 2011 LEHIGH U N I V E R S I T Y x = 0.1 x = 0.2 x = 0.3 x = 0.4 x = 0.5 x = 0.6 x = 0.7 x = 0.8 x = 0.9 radii Simultaneous control of the  (x) profile and  N (shot # : control t = 1.5 s) 9  N control  -profile control  (x= 0.1, 0.2, …0.9) & targets (lines) Time

D. Moreau, ITPA-IOS Meeting, Kyoto, October 18-21, 2011 LEHIGH U N I V E R S I T Y Simultaneous control of the  (x) profile and  N (shot # : control t = 1.5 s) 10 5 actuators (no saturation) Vsurf actuator control & Ip Cost function minimization

D. Moreau, ITPA-IOS Meeting, Kyoto, October 18-21, 2011 LEHIGH U N I V E R S I T Y MHD Ip Simultaneous control of the  (x) profile and  N shot # : control t = 1 s (ramp-up) 11  N control  -profile control x = 0.1 x = 0.2 x = 0.3 x = 0.4 x = 0.5 x = 0.6 x = 0.7 x = 0.8 x = 0.9 radii  (x= 0.1, 0.2, …0.9) & targets (lines) Time

D. Moreau, ITPA-IOS Meeting, Kyoto, October 18-21, 2011 LEHIGH U N I V E R S I T Y Simultaneous control of the  (x) profile and  N shot # : control t = 1 s (ramp-up) 12 5 actuators (MHD  NB-Bal saturation) Vsurf actuator control & Ip Cost function minimization

D. Moreau, ITPA-IOS Meeting, Kyoto, October 18-21, 2011 LEHIGH U N I V E R S I T Y 13 Lehigh University Approach to Feedback Control Design: Data-driven Model-based Current Profile Control –Both static and dynamic control-oriented plasma response models are embedded in the control synthesis. –A cost functional is defined to quantify the control objectives: Tracking error minimization Disturbance rejection Control power minimization –Stabilizing controllers, which are robust against model uncertainties, are synthesized by minimizing different norms (H ∞ and H 2 norms) of the cost functional subject to the control-oriented model. –These controllers do not need PID-like empirical tuning. –The controllers are augmented with model-based anti-windup compensators to overcome detriments effects due to actuator saturation.

D. Moreau, ITPA-IOS Meeting, Kyoto, October 18-21, 2011 LEHIGH U N I V E R S I T Y 14 Data-driven Model-based Control of  (x) =1/q(x) –In the DIII-D PCS different magnetic profiles (  (  ),  (  ), q(  ) or  (  )=  /  ) can now be obtained in real time from a complete equilibrium reconstruction using data from the MSE diagnostic. –Figure on the left illustrates both simulated and experimental evolutions for the rotational transform  (  ) at normalized radii  =0.2, 0.4, 0.5, 0.6, 0.8 for shot –Tracking of the  (  ) target profile was achieved by regulating: Total plasma current Co-injection and balanced beam powers (counter-injection beam not available) Total ECH/ECCD power –Artificial input disturbances were introduced at t=3.5 sec.

D. Moreau, ITPA-IOS Meeting, Kyoto, October 18-21, 2011 LEHIGH U N I V E R S I T Y 15 Data-driven Model-based Control of  (x) =1/q(x) Tight regulation is achieved before t=3.5 sec. A significant tracking error in the inner part of the profile is noted after the artificial disturbance is introduced at t=3.5 sec. The controller reacts by decreasing the tracking error at the inner point and slightly increasing the error at the outer points with the ultimate goal of improving the overall profile tracking. Ip regulation was poor due to setup problem. More control authority is expected with stronger Ip regulation and counter injection beam.

D. Moreau, ITPA-IOS Meeting, Kyoto, October 18-21, 2011 LEHIGH U N I V E R S I T Y Control of  (x),  (x) = 1/q(x) and combined control of  (x) and  N have been achieved for the first time on DIII-D using either 4 or 5 actuators : Co-NBI, (Cnt-NBI), Bal-NBI, ECCD, Vsurf New PCS with profile control algorithm was qualified and worked perfectly  (x) and  N control was switched on during current ramp-up. Combined feedback control of  (x) and  N was successful up to [1s-6s] Proposal for ITPA-IOS 2012 : Integrated magnetic and kinetic plasma control More experiments on DIII-D with different target current profiles and  N. Add control of the rotation and/or Ti profiles (real-time CER) Start profile control experiments on other devices (Tore Supra, TCV …?) Summary and proposal for 2012 … 16